The central and non-central matrix variate Dirichlet type III distribution : theory and methods
René Ehlers, Andriëtte Bekker, J. J. J. Roux
Abstract
René Ehlers, Andriëtte Bekker, J. J. J. Roux
Abstract
The methods of multivariate analysis depend upon the multivariate normal distribution. If a sample is drawn from a multivariate normal distribution, the sample covariance matrix has a Wishart distribution. Therefore, this distribution plays an important role in inferential procedures. The matrix variate Beta distribution is closely connected to the Wishart distribution.
OpenAlex reports 2 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
The methods of multivariate analysis depend upon the multivariate normal distribution. If a sample is drawn from a multivariate normal distribution, the sample covariance matrix has a Wishart distribution. Therefore, this distribution plays an important role in inferential procedures. The matrix variate Beta distribution is closely connected to the Wishart distribution.
Key concepts: Wishart distribution, Matrix t-distribution, Matrix normal distribution, Inverse-Wishart distribution, Mathematics, Normal-Wishart distribution, Random variate, Scatter matrix